We propose new solid-state rotational-echo double resonance (REDOR) and Double REDOR NMR experiments to characterize the structures of peptides, proteins, or DNA complexed or aggregated with cell walls or membranes. Many of these complexes will be examined in situ. All of the systems are heterogeneous, insoluble, and not suited to analysis by diffraction or solution-state NMR methods. Our goal is not the determination of total structure, but rather the analysis of restricted regions (interfaces, channels, binding sites) important to biological function. Our approach uses (i) REDOR methods for internuclear distances as great as 15 Angstrom units; (ii) specific stable-isotope labeling schemes for three and four different kinds of nuclei; and (iii) NMR probes tuned simultaneously to four, five, and even six radiofrequencies. In the last grant period, we examined (i) peptidoglycans in bacterial cell walls; (ii) antimicrobials in multilamellar vesicles; and (iii) antifungals in oriented phospholipid bilayers. As a part of this program, we have also developed analytical methods for interpreting REDOR experiments of multi-spin systems. Based on this work, we are now in position to extend our solid-state NMR approach to cell-wall and membrane problems of immediate biomedical interest. We will use REDOR to characterize the inhibiting complexes of vancomycin with staphylococcus aureus cell-wall precursors in situ to determine (i) the mode of action of vancomycin and vancomycin derivatives. Some anti- fungals are believed to work by aggregation leading to the formation of ion channels or pores. REDOR will detect directly aggregates of antifungals in multilamellar vesicles which will lead to our characterization of (ii) bilayer pore formation by magainins and amphotericin B. We intend to develop the methodology needed to extend REDOR to the broad class of ligand-activated G-protein coupled receptors and use this technology to determine (iii) the conformation of a G- protein fragment bound to rhodopsin. Finally, we are working with surface-active particles that complex DNA and have potential as transfection vehicles in gene therapy. Key design problems are to optimize (iv) DNA compaction and transfection by amphiphilic particles, which we believe can be solved using insights from REDOR. Because stable free radicals having magnetic properties that are ideal for dynamic nuclear polarization can be buried in the hydrophilic core of these particles, new types of electron-based REDOR experiments will be developed to increase the range of quantitative distance measurements in biological solids by an order of magnitude.